Private Prisons in the

United States

Executive Summary

Cambridge, MA
Lexington, MA
Hadley, MA
Bethesda, MD
Washington, DC
Chicago, IL
Cairo, Egypt July 16, 1998
Johannesburg, South Africa

Prepared by
Douglas McDonald, Ph.D.
Elizabeth Fournier
Malcolm Russell-Einhourn, J.D.
Stephen Crawford
Abt Associates Inc.
55 Wheeler Street With the assistance of:
Cambridge, MA 02138 Julianne Nelson, Ph.D.
Gerald G. Gaes, Ph.D.
Scott D. Camp, Ph.D.
William G. Saylor

Abt Associates Inc.

Summary
This report examines the current state of practice, law, and research with respect to privately operated
prisons in the United States, at all levels of security. The body of current law relevant to privately
operated prisons is reviewed, including federal and state statutes, regulations, and case law. Current
knowledge regarding the cost and cost-effectiveness of privately operated prisons is assessed by
means of reviews and evaluations of existing research on cost and quality of performance.

The report’s focus is limited principally to contracting by state and federal governments for the
private management and operation of facilities that are analogous to secure state or federal prisons.
Although the private imprisonment industry also provides secure detention for defendants awaiting
trial, for illegal immigrants, for prisoners transitioning back into their community, and for juveniles,
such facilities are not examined here. Contracting for these latter facilities was excluded from this
study for several reasons. First, state and federal correctional agencies are responsible for sending
most of the prisoners held in privately operated facilities&about 80 percent of all privately held
prisoners at year-end 1997. Second, most of the studies comparing cost and performance of privately
and publicly operated facilities examine state prisons or their private equivalents, rather than local
jails and privately operated detention facilities. To survey more than 3,000 local governments to
identify their practices of contracting for privately operated detention would have required more time
and resources than were available for this study.

By 1997, the numbers of adults in privately operated secure facilities of all sorts&prisons, jails, and
illegal immigrant detention centers&reached about 64,000, which constituted less than three percent
of the entire United States population of confined adults. The number of facilities had grown to
approximately 140, and the industry’s revenues probably approached $1 billion.

Information about the subset of the industry that encompasses state and federal contracting for secure
imprisonment was obtained from a variety of sources, including a survey of contracting practices
conducted by Abt Associates Inc. of prison administrators in all state and federal governments, in the
District of Columbia, Puerto Rico, and the Virgin Islands. All but two jurisdictions provided
information. This survey determined that 28 of these jurisdictions had 91 active contracts at year-end
1997 with 84 different privately operated facilities. These facilities reportedly held 37,651 prisoners
at that time. The surveyed jurisdictions also reported having an additional 14,719 prisoners in still
other privately operated facilities. Many of these prisoners had been transferred to another
jurisdiction under the authority of an intergovernmental agreement, and the receiving government had
placed them in a privately operated facility with which it had a contract. Others were held in non-
secure facilities with which the governments contracted.

Ten correctional agencies were responsible for placing the vast majority of prisoners in privately
operated facilities. These were, in descending order, the Federal Bureau of Prisons, the States of
Texas, Florida, Oklahoma, Louisiana, Tennessee, California, Mississippi, and Colorado, and the
District of Columbia. The most commonly reported reason for contracting with private management
was not to reduce costs but to alleviate overcrowding in the public system and to acquire needed beds
quickly. It is not surprising, therefore, that most (sixty percent) of the privately operated facilities are
privately owned. In most places, contracting for imprisonment services was not taken at the initiative

Abt Associates Inc. Private Prisons iii

of the correctional agency, but was instead mandated by either the legislature or the chief executive of
the jurisdiction, typically the governor.

Does Contracting for Prison Operations Save Money?

Even though the quest for financial savings was not given by prison administrators as their primary
reason for contracting, it is no doubt a common hope among policymakers that governments will save
money by doing so. Some proponents argue that evidence exists of substantial savings as a result of
privatization. Indeed, one asserts that a typical American jurisdiction can obtain economies in the
range of 10 to 20 percent. Our analysis of the existing data does not support such an optimistic view.

Comparing public and private prisons’ costs is complicated for a variety of reasons. Comparable
public facilities may not exist in the same jurisdiction. Private facilities may differ substantially from
other government facilities in their functions (e.g., the private facility in Arizona houses men and
women, or some in Texas that are used for drug abuse treatment services or for pre-release
populations placed in halfway houses by other jurisdictions). Or they may differ in their age, design,
or the security needs of inmates housed, all of which affect the cost of staffing them. Cost
comparisons are also difficult because public and private accounting systems were designed for
different purposes; that is, public systems were not designed principally for cost accounting.
Spending to support imprisonment is often borne and reported by agencies other than the correctional
department, and computation of these costs is often difficult for lack of data. The annual costs of
"using up" the physical assets are not counted in the public sector, as capital expenditures are
generally valued only in the year that they are made, rather than being spread across the life of the
assets. Nor is the cost to the taxpayer of contracting readily apparent from tallies of payments to
contractors. Governments incur expenses for contract procurement, administration, and monitoring;
for medical costs above amounts capped by contracts; and for sentence computation, transportation,
and other activities performed by governments. Cost comparisons often fail to account for such
expenditures.

Of the approximately 140 secure confinement facilities currently in existence, or the 84 that held
active contracts with state or federal agencies at the end of 1997, only a handful have been studied to
learn if contracting is less costly to the taxpayer. Fewer still have employed reasonably strong
research designs and reported the data in sufficient detail to permit an assessment of the validity of the
findings. The results are mixed and subject to different interpretations.

Some studies report finding that contracting saved the taxpayers money; others report small
differences, if any. Some of these apparent differences may not reflect actual savings but may instead
be accounting artifacts, especially those associated with lower estimated costs of government
overhead activities. That is, signing lower overhead costs to privately operated facilities may not
result in actual savings because these costs are quite fixed. Savings will accrue to taxpayers only if
(a) overhead activities are cut back as a result of contracting (an unlikely event where only a few
facilities are contracted in a larger public system), or (b) if planned increases in overhead costs are
averted as a result of contracting. It is unclear in the studies reviewed if purported savings associated
with reduced use of government overhead functions actually resulted in less spending.

Abt Associates Inc. Private Prisons iv

Other apparent sources of savings in some states reflect lower spending for prisoner health care and,
perhaps, in other aspects of facility operations, including lower salaries for line staff in some
jurisdictions. In some states, payments to contractors may be offset somewhat by accounting
deductions or adjustments for contractors' tax payments to governments.

Our conclusion regarding costs and savings is that the few existing studies and other available data do
not provide strong evidence of any general pattern. Some states may be willing to pay high prices for
private imprisonment if they need the beds to solve short-term deficiencies. In other states,
expenditures for contracting may indeed be lower than for direct public provision. However, the
bottom line with respect to costs and savings is difficult to discern given the data and the assumptions
made by the analysts. Drawing conclusions about the inherent superiority of one or the other mode of
provision, based on a few studies, is premature.

Do Privately Operated Facilities Provide Better Services?

Interest in better imprisonment services was infrequently given as a reason for contracting, but the
issue of whether governments or private firms seek lower costs by sacrificing service quality is a
major concern, especially in an environment where public safety and staff and inmate lives are at risk,
and where lawsuits are common. Unfortunately, assessing the quality of imprisonment services is
controversial, in part because the various objectives government and the larger society establish for
prisons do not always suggest unambiguous measures. At a minimum, we expect prisons to meet the
standards established by the courts.

Most contracts require that privately operated facilities conform to the law, rules, and regulations that
prevail in the public correctional agency of that jurisdiction. Performance objectives are thereby
framed in procedural terms, rather than outcomes to be achieved by means of imprisonment. Contract
monitoring is highly variable from one jurisdiction to another, with about half of the contracts
involving daily on-site monitoring by government officials.

Few studies have been conducted to compare the relative performance of privately and publicly
operated prisons. Most are affected by a variety of methodological problems that severely limit the
conclusions that one can draw from them. These include insufficiently comparable public facilities
and insufficient attention to the possibility that observed differences resulted from dissimilar
populations of prisoners or dissimilar facility designs. Given these shortcomings and the paucity of
systematic comparisons, one cannot conclude whether the performance of privately managed prisons
is different from or similar to that of publicly operated ones.

With respect to public safety and inmate programming, the available data do not support definite
conclusions. Programs for inmates are especially important in prisons because they prepare offenders
to become reintegrated into society upon their release. Well designed and implemented programs
increase the likelihood that prisoners will become law-abiding citizens. The available surveys of
either privately or publicly operated facilities do not provide the information needed to compare the
quality of such programs or the extent of prisoners’ engagement with them.

Abt Associates Inc. Private Prisons v

Legal Issues Relevant to Contracting for Imprisonment

Whereas the legality of governments delegating correctional authority to private firms was much
debated in the 1980s, it now appears that objections to prison privatization on constitutional
delegation grounds have little force. Unless a government has absolutely no persuasive statutory
authority for entering into private prison contracts, courts will be reluctant to invalidate contacts on
delegation grounds. Only delegated rulemaking and adjudication functions that directly purport to
exercise a government power are deemed to require special constitutional due process safeguards and
to be subject to heightened judicial scrutiny. No clear case law has been developed to define with
precision how general due process standards will be applied to private prisons. These issues have
been handled in some states through legislation, and through contracts in others.

With respect to liability, private prisons are generally treated as "state actors" for purposes of civil
rights suits, so that all relevant constitutional requirements apply with equal force to private as well as
public facilities. Private prison employees do not have the advantage of "qualified immunity" shields,
nor are they protected by other governmental immunities that limit monetary damages available to
inmates suing over prison conditions. Governments cannot shield themselves entirely from liability
by means of contracting, but they can lessen their exposure by doing so. In the vast majority of suits
brought by prisoners, alleging individual rather than systemic group harms, governments will
generally not be deemed to have knowledge of the specific acts and injuries, and will not be held
responsible. Public correctional authority’s litigation costs can be kept as low as possible by requiring
that a private contractor indemnify them and have relevant public entities named as an insured on the
contractor’s comprehensive general liability insurance policy.

While the right to strike exists in privately operated prisons, the risk of such disruptions can be
minimized by private firms seeking to have employees agree, individually or collectively, to no-strike
pledges. Moreover, contractors can also insist on notification requirements that allow them to make
arrangements for assumption of certain essential responsibilities in the event of a labor action.

The possibility of bankruptcy has excited much concern regarding private imprisonment, but this has
not yet been a problem in the industry. The few exceptions involved some small firms that
speculated by building facilities in the absence of contracts with an agency. Public correctional
agencies should nonetheless seek to protect themselves against the untoward consequences of
bankruptcy by means of proper monitoring and careful contracting.

Regarding the use of force, including deadly force: the major legal issue is whether the use of force is
properly mandated by the relevant laws of the jurisdiction. Without proper enabling legislation or
contractual provisions authorizing the use of force by designated private prison officials, it is possible
that such persons and their firms could face civil and criminal liability.

Sending prisoners from one jurisdiction to privately operated prisons in another poses some special
legal questions. Several jurisdictions rely not on contracts but instead on intermediary governments
to place prisoners in private facilities. Several jurisdictions also lease bed space directly from private
contractors in foreign jurisdictions, sometimes in prisons that are built and operated on a speculative
basis. These interjurisdictional arrangements have raised questions about the sufficiency of legal
authority and regulation. For example, receiving facilities may lack legal authority to allow them to

Abt Associates Inc. Private Prisons vi

respond to a variety of critical situations, including escapes and other prison emergencies. Legislation
may be needed to govern these practices.

The success or failure of a private prison arrangement may depend upon the skill with which contracts
are designed and negotiated. Public authorities must give close attention to the purposes served by
contracting and the degree of specificity that they seek to build into the agreement.

Implications for Federal Prisons

Bureau of Prisons officials assert that the private sector’s experience in operating higher security
prisons or managing inmate populations with higher security needs is too limited to warrant the
privatization of such facilities in the federal system. (Correctional authorities distinguish between the
physical security of prisons and and the security needs of inmates.) To be sure, maximum security
facilities and prisoners remain almost entirely in the government-operated sector: only four percent
(1,661) of prisoners in privately operated facilities were classified as maximum security, whereas 20
percent (or 200,000) of all prisoners in government facilities were so classified.

The private sector’s experience with what the Federal Bureau of Prisons terms medium security
prisoners is more ambiguous, however. Surveyed correctional agencies reported that half of their
prisoners housed in private facilities under contract at year-end 1997 were classified as medium
security. But procedures for classifying prisoners according to their security needs vary from one
jurisdiction to the next. It is not clear that a prisoner classified as medium security in one will be
classified the same in all others. For example, a comparison of prisoners classified by Louisiana and
the Federal Bureau of Prisons suggests that the private sector’s experience in at least this one state
may not be comparable to the federal government’s practices at the medium security level.

Eighty percent of all prisoners in two privately operated facilities and in a third government operated
facility in Louisiana were classified by the state as being medium security; the remaining prisoners
were split evenly between maximum and minimum security. A random sample of prisoners from
these three facilities was picked and they were classified according to federal procedures, which uses
a four- rather than a three-tiered classification system: high, medium, low, and minimum security.
Approximately 33 percent of these prisoners would have been classified as medium-security in
federal prisons; 29 percent and 28 percent would have been classified as "low" and "minimum"
security, respectively. The proportion of high security prisoners was the same as classified by
Louisiana (10 percent). This distribution of prisoners so classified is not characteristic of medium-
security prisons in the federal prison system. Rather, it is closer to that found in low-security prisons.
Lacking similar data from other states, it is not possible to determine how the security needs of
inmates housed in other privately operated facilities correspond to those prisoners held in federal
facilities.

Abt Associates Inc. Private Prisons vii